抄録
This article examines the size distribution of “local infection clusters” (“clusters” in short; see text for the definition) of measles and rubella from 2008 to 2013 in Japan. When the logarithm of cluster sizes were plotted on the x-axis and the logarithm of their frequencies in the y-axis, the plots fell on a rightward descending straight line. The size distribution was found to follow power law. As the size distribution of the clusters could be equated with that of local secondary infections initiated by one patient, the size distribution of the clusters was in fact that of the “effective reproduction numbers” in the local level. As the power law distribution had no typical sizes, it was suggested that measles or rubella epidemic in Japan had no typical reproduction number. The higher was the population size and the higher was the total number of patients, the flatter was the slope of the plots, i.e., the larger was the proportion of larger clusters. Epidemic of measles or rubella in Japan could be represented more appropriately by the cluster size frequency distribution rather than by reproduction number.
